git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[JGIT PATCH 10/10] BROKEN TEST: ObjectLoader stays valid across repacks

From
Shawn O. Pearce <spearce@spearce.org>
Date
Apr 21, 2009, 01:21 UTC
Message-ID
<1240276872-17893-11-git-send-email-spearce@spearce.org>
In-Reply-To
<1240276872-17893-10-git-send-email-spearce@spearce.org>
This doesn't doesn't work.
What we are trying to verify is that an ObjectLoader remains valid
if the underlying storage for the object has moved, such as when a
repository is repacked, the old pack was deleted, and the object is
now in the new pack.
---
 .../org/spearce/jgit/lib/ConcurrentRepackTest.java |   46 ++++++++++++++++++++
 1 files changed, 46 insertions(+), 0 deletions(-)
diff --git a/org.spearce.jgit.test/tst/org/spearce/jgit/lib/ConcurrentRepackTest.java b/org.spearce.jgit.test/tst/org/spearce/jgit/lib/ConcurrentRepackTest.java
index 6eb368c..b1b1af2 100644
--- a/org.spearce.jgit.test/tst/org/spearce/jgit/lib/ConcurrentRepackTest.java
+++ b/org.spearce.jgit.test/tst/org/spearce/jgit/lib/ConcurrentRepackTest.java
@@ -41,6 +41,7 @@
 import java.io.File;
 import java.io.FileOutputStream;
 import java.io.IOException;
+import java.util.Arrays;
 
 import org.spearce.jgit.errors.IncorrectObjectTypeException;
 import org.spearce.jgit.errors.MissingObjectException;
@@ -128,6 +129,51 @@ public void testObjectMovedWithinPack()
 		assertEquals(o2.name(), parse(o2).name());
 	}
 
+	public void testObjectMovedToNewPack2()
+			throws IncorrectObjectTypeException, IOException {
+		// Create an object and pack it. Then remove that pack and put the
+		// object into a different pack file, with some other object. We
+		// still should be able to access the objects.
+		//
+		final Repository eden = createNewEmptyRepo();
+		final RevObject o1 = writeBlob(eden, "o1");
+		final File[] out1 = pack(eden, o1);
+		assertEquals(o1.name(), parse(o1).name());
+
+		final ObjectLoader load1 = db.openBlob(o1);
+		assertNotNull(load1);
+
+		final RevObject o2 = writeBlob(eden, "o2");
+		pack(eden, o2, o1);
+
+		// Force close, and then delete, the old pack.
+		//
+		whackCache();
+		delete(out1);
+
+		// Now here is the interesting thing... can the loader we made
+		// earlier still resolve the object, even though its underlying
+		// pack is gone, but the object still exists.
+		//
+		final ObjectLoader load2 = db.openBlob(o1);
+		assertNotNull(load2);
+		assertNotSame(load1, load2);
+
+		// Currently load1.getCachedBytes NPEs due to the openCount == 1,
+		// but fd == null. Aside from the underlying pack being gone, the
+		// WindowCache didn't reset the openCount to 0 when it closed the
+		// file during the mass cache eviction. Thus we never tried to do
+		// an subsequent open.
+		//
+		final byte[] data2 = load2.getCachedBytes();
+		final byte[] data1 = load1.getCachedBytes();
+		assertNotNull(data2);
+		assertNotNull(data1);
+		assertNotSame(data1, data2); // cache should be per-pack, not per object
+		assertTrue(Arrays.equals(data1, data2));
+		assertEquals(load2.getType(), load1.getType());
+	}
+
 	private static void whackCache() {
 		final WindowCacheConfig config = new WindowCacheConfig();
 
-- 
1.6.3.rc1.188.ga02b
Previous: Shawn O. PearceNext: Robin Rosenberg
Message 11 of 17 in “Object access improvements”
  1. 00/10 Object access improvementsShawn O. Pearce, Apr 21, 2009
  2. 01/10 Safely handle closing an already closed WindowedFileShawn O. Pearce, Apr 21, 2009
  3. 02/10 Change empty tree test case to use a temporary repositoryShawn O. Pearce, Apr 21, 2009
  4. 03/10 Replace hand-coded read fully loop with NB.readFullyShawn O. Pearce, Apr 21, 2009
  5. 04/10 Clear the reverse index when closing a PackFileShawn O. Pearce, Apr 21, 2009
  6. 05/10 Introduce a new exception type for an in-place pack modificationShawn O. Pearce, Apr 21, 2009
  7. 06/10 Refactor object database access with new abstractionShawn O. Pearce, Apr 21, 2009
  8. 07/10 Rescan packs if a pack is modified in-place (part 1)Shawn O. Pearce, Apr 21, 2009
  9. 08/10 Scan for new packs if GIT_DIR/objects/pack has been modifiedShawn O. Pearce, Apr 21, 2009
  10. 09/10 Add test cases for loading new (or replaced) pack filesShawn O. Pearce, Apr 21, 2009
  11. 10/10 BROKEN TEST: ObjectLoader stays valid across repacksShawn O. Pearce, Apr 21, 2009
  12. Robin RosenbergApr 21, 2009
  13. Shawn O. PearceApr 22, 2009
  14. Shawn O. PearceApr 22, 2009
  15. 09/10 Add test cases for loading new (or replaced) pack filesShawn O. Pearce, Apr 21, 2009
  16. Shawn O. PearceApr 21, 2009
  17. 08/10 Scan for new packs if GIT_DIR/objects/pack has been modifiedShawn O. Pearce, Apr 21, 2009

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.